Prompting Large Vision-Language Models for Compositional Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913202580750336 |
|---|---|
| author | Ossowski, Timothy Jiang, Ming Hu, Junjie |
| author_facet | Ossowski, Timothy Jiang, Ming Hu, Junjie |
| contents | Vision-language models such as CLIP have shown impressive capabilities in encoding texts and images into aligned embeddings, enabling the retrieval of multimodal data in a shared embedding space. However, these embedding-based models still face challenges in effectively matching images and texts with similar visio-linguistic compositionality, as evidenced by their performance on the recent Winoground dataset. In this paper, we argue that this limitation stems from two factors: the use of single vector representations for complex multimodal data, and the absence of step-by-step reasoning in these embedding-based methods. To address this issue, we make an exploratory step using a novel generative method that prompts large vision-language models (e.g., GPT-4) to depict images and perform compositional reasoning. Our method outperforms other embedding-based methods on the Winoground dataset, and obtains further improvement of up to 10% accuracy when enhanced with the optimal description. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_11337 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Prompting Large Vision-Language Models for Compositional Reasoning Ossowski, Timothy Jiang, Ming Hu, Junjie Computer Vision and Pattern Recognition Artificial Intelligence Vision-language models such as CLIP have shown impressive capabilities in encoding texts and images into aligned embeddings, enabling the retrieval of multimodal data in a shared embedding space. However, these embedding-based models still face challenges in effectively matching images and texts with similar visio-linguistic compositionality, as evidenced by their performance on the recent Winoground dataset. In this paper, we argue that this limitation stems from two factors: the use of single vector representations for complex multimodal data, and the absence of step-by-step reasoning in these embedding-based methods. To address this issue, we make an exploratory step using a novel generative method that prompts large vision-language models (e.g., GPT-4) to depict images and perform compositional reasoning. Our method outperforms other embedding-based methods on the Winoground dataset, and obtains further improvement of up to 10% accuracy when enhanced with the optimal description. |
| title | Prompting Large Vision-Language Models for Compositional Reasoning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2401.11337 |